JupiteX Get the app
Defence & Security19 Nov 2026 · about 7 min

Unauthenticated remote code execution in Nginx

The brief

NGINX is software used to receive and route traffic to production applications and APIs. It can act as a shared gateway between incoming requests and the services that handle them. That makes it important infrastructure: a weakness in NGINX itself can expose websites even when their application code is secure. For example, one NGINX deployment may handle traffic for several applications. If attackers take control of that server, they may be able to execute code, steal sensitive data, tamper with websites, or disrupt services. The article reports that testing of one flaw achieved code execution on NGINX serving ordinary static files, without a backend application or third-party modules. Both reported vulnerabilities require HTTP/2 to be enabled on the affected server. The researchers reported them to F5, which coordinated remediation and public disclosure. Organizations are advised to identify affected deployments and update to the latest release containing multiple security fixes. The central risk is that shared traffic infrastructure can widen the impact of a single vulnerable server.

01

What is NGINX, and what does it do for websites and APIs?

NGINX is software used to receive and route traffic to production applications and APIs. It can act as a shared gateway between incoming requests and the services that handle them. That makes it important infrastructure: a weakness in NGINX itself can expose websites even when their application code is secure.

For example, one NGINX deployment may handle traffic for several applications. If attackers take control of that server, they may be able to execute code, steal sensitive data, tamper with websites, or disrupt services. The article reports that testing of one flaw achieved code execution on NGINX serving ordinary static files, without a backend application or third-party modules.

Both reported vulnerabilities require HTTP/2 to be enabled on the affected server. The researchers reported them to F5, which coordinated remediation and public disclosure. Organizations are advised to identify affected deployments and update to the latest release containing multiple security fixes. The central risk is that shared traffic infrastructure can widen the impact of a single vulnerable server.

02

What does unauthenticated remote code execution mean for an attacker and the server?

Unauthenticated remote code execution means an attacker can make a vulnerable server run their chosen code over a network without first proving an authorized identity. In these cases, the article describes two critical vulnerabilities in NGINX and confirms that each could allow an unauthenticated attacker to execute code on the server.

If the attacker succeeds, the NGINX server may be compromised. The article lists possible consequences: stealing sensitive data, tampering with websites, or disrupting services. Because NGINX can route traffic for several applications, taking over one server may also put those services at risk. The article says successful exploitation could let attackers take control of affected NGINX servers.

Both flaws require HTTP/2 to be enabled, but the article does not give a complete list of conditions for exploitation. F5 issued a coordinated security release on November 18, 2026, and organizations should identify affected deployments and apply the security updates. The risk is serious because the flaws affect the traffic-handling software itself, even where application code is secure.

03

How many critical flaws were found, and how far back did the oldest one date?

The researchers uncovered two critical remote code execution vulnerabilities in NGINX: CVE-2026-123633 and CVE-2026-123007. Both received a CVSS 4.0 score of 9.2. The vulnerabilities were independently uncovered by Wiz Research, Ikotas Labs, Metnir, Rapid7, and Xint, and were reported to F5, the company behind NGINX.

The older flaw, CVE-2026-123007, affects versions dating back to NGINX 1.9.5 in 2015. The article identifies it as an integer underflow vulnerability. CVE-2026-123633 is a use-after-free vulnerability that had been present for nearly six years. Both can allow unauthenticated attackers to execute code, subject to the conditions described in the report.

F5 coordinated disclosure and released security fixes on November 18, 2026. The article recommends that organizations identify affected deployments and update to the latest release, which includes multiple security fixes. It also notes that AI-assisted research helped uncover the flaws, illustrating how longstanding memory-safety bugs can become discoverable after remaining hidden for years.

04

What must be enabled on an affected NGINX server for these vulnerabilities to be exploitable?

The article identifies one shared requirement for the two NGINX vulnerabilities: HTTP/2 must be enabled on the affected server. HTTP/2 is a protocol for communication between clients and servers, but the article does not explain the vulnerabilities’ detailed technical triggers. It also cautions that additional conditions apply.

For example, an organization running NGINX with HTTP/2 enabled should treat that deployment as potentially exposed and check whether it is affected. The article’s testing of CVE-2026-123007 demonstrated code execution on NGINX serving ordinary static files, without a backend application or third-party modules. This shows that exposure is not limited to servers running a particular application setup.

The source does not state that disabling HTTP/2 is a complete or recommended fix, nor does it provide all exploitation conditions. F5 released security updates on November 18, 2026. Organizations are advised to identify affected deployments and update to the latest release with multiple security fixes, rather than assume their application code prevents the NGINX-level flaw.

05

Why can compromising one NGINX server put several websites or APIs at risk?

NGINX often handles traffic for several applications. It receives requests and routes them to the production applications and APIs they are intended to reach. When that shared traffic-handling server is compromised, attackers may gain a position that affects multiple services, rather than just one application.

The article says successful exploitation could let attackers execute code on the affected server, steal sensitive data, tamper with websites, or disrupt services. It also reports that the vulnerabilities are in NGINX itself. As a result, websites and APIs can be exposed even when their own application code is secure. Testing of CVE-2026-123007 demonstrated code execution with ordinary static files and no backend application or third-party modules.

Both vulnerabilities require HTTP/2 to be enabled, and organizations should identify affected NGINX deployments. F5 coordinated disclosure and released security updates on November 18, 2026. The article recommends updating to the latest release with multiple security fixes. The shared role of NGINX makes reducing exposure important across all services that rely on an affected deployment.

06

What are a use-after-free and an integer underflow, and how can they lead to memory errors?

The article identifies CVE-2026-123633 as a use-after-free and CVE-2026-123007 as an integer underflow. It does not explain their internal memory operations. In general, a use-after-free occurs when software continues to access a memory location after releasing it. That location may have been reused, so the stale access can read or change unexpected data.

An integer underflow is a general arithmetic error: subtracting more than a value can represent may make the result wrap to a very large value, depending on the number type. If software uses that result to check sizes or manage memory, incorrect decisions can follow. These are general explanations, not a description of the exact code paths in the reported NGINX flaws.

The article says both flaws can enable unauthenticated code execution when HTTP/2 is enabled, and CVE-2026-123007 dates back to NGINX 1.9.5 in 2015. It describes such memory-safety flaws as difficult to discover because they can involve subtle interactions across memory operations and code paths. F5 released security updates, and organizations should update affected deployments.

07

How does HTTP/2 flow control manage data, and why can incorrect accounting create a security flaw?

The article states that HTTP/2 must be enabled for both NGINX vulnerabilities to be exploitable, but it does not describe HTTP/2 flow control or identify it as the cause. In general, HTTP/2 flow control helps prevent a sender from overwhelming a receiver with data. It tracks how much data may be sent within a connection and within individual streams.

In typical HTTP/2 flow control, a receiver grants a data window and can increase that allowance as it processes data. If software tracks the data or window incorrectly, it may make unsafe assumptions about how much data is available or how much has been handled. Depending on implementation, that kind of accounting error could contribute to memory or bounds problems. This is general background, not a claimed mechanism for these NGINX vulnerabilities.

The article identifies one flaw as a use-after-free and the other as an integer underflow, but does not connect either one to a specific flow-control bug. Both are critical, unauthenticated remote code execution vulnerabilities, and both require HTTP/2 to be enabled. F5 released security fixes on November 18, 2026; affected deployments should be updated.

This brief was written by AI from the original reporting and checked by other models. Names, figures and quotes come from the source; read it for full context.

Read more in the JupiteX app

Pulse is free. New stories every 4 hours, each one broken into the questions that explain it.

Or read more news on the web